Resource MEMORY_MB Unable to retrieve providers information

[Victor Serbu]

Hello
When I click on admin -> compute -> hypervisor I get error and nothing is displayed in resource providers summary.
This
This happens because i have an environment in which i have activate pci passthrough in nova and there are tracked in placement: https://docs.openstack.org/nova/2023.1/admin/pci-passthrough.html#pci-tracking-in-placement

and this inventory doesn't have MEMORY_MB or DISK_GB (i looked at https://opendev.org/openstack/horizon/src/branch/master/openstack_dashboard/api/placement.py#L117-L127)

eg
one compute node where i have activated pci passthrough
```
(openstack) [osc@ansible-3 ~]$ openstack resource provider show 1fe9d32f-43cd-445a-8a49-a68f9ff5158f
+----------------------+--------------------------------------+
| Field | Value |
+----------------------+--------------------------------------+
| uuid | 1fe9d32f-43cd-445a-8a49-a68f9ff5158f |
| name | compute-19_0000:04:00.0 |
| generation | 44 |
| root_provider_uuid | e9cb6a8d-e638-4245-bf79-981211c5a232 |
| parent_provider_uuid | e9cb6a8d-e638-4245-bf79-981211c5a232 |
+----------------------+--------------------------------------+
(openstack) [osc@ansible-3 ~]$ openstack resource provider usage show 1fe9d32f-43cd-445a-8a49-a68f9ff5158f
+----------------------+-------+
| resource_class | usage |
+----------------------+-------+
| CUSTOM_PCI_10DE_2330 | 1 |
+----------------------+-------+
```

One simple compute node show as
```
(openstack) [osc@ansible-3 ~]$ openstack resource provider usage show f7af998e-1563-4a55-9145-4ee5f527d12b
+----------------+--------+
| resource_class | usage |
+----------------+--------+
| VCPU | 412 |
| MEMORY_MB | 824320 |
| DISK_GB | 0 |
+----------------+--------+

```

[Andrew Bonney (andrewbonney)]

We're seeing the same issue.

[Sam Schmitt (samcat116)]

I see the same thing with Ironic baremetal nodes tracked in placement.